Settings
Input parameters
The following set of tabs collects the parameters used to perform the comorbidity analysis:

Comorbidity analysis parameters

Patient Data input settings

  • Data full path:  /home/ronzano/Eclipse_WS/comorbidity4j-github/example/input/
  • Patient Data file name:  patientData.csv
  • Column Name of patientData file, column patient_id:  patient_id
  • Column Name of patientData file, column patient_dateBirth:  patient_birth_date
  • Column Name of patientData file, column patient_sex:  patient_sex
  • Column Name of patientData file, column patient_facet_1:  Race

Admission Data input settings

  • Data full path:  /home/ronzano/Eclipse_WS/comorbidity4j-github/example/input/
  • Aadmission Data file name:  admissionData.csv
  • Column Name of admissionData file, column patient_id:  patient_id
  • Column Name of admissionData file, column admission_id:  admission_id
  • Column Name of admissionData file, column admissionStartDate:  admission_date

Diagnosis Data input settings

  • Data full path:  /home/ronzano/Eclipse_WS/comorbidity4j-github/example/input/
  • Diagnosis Data file name:  diagnosisData.csv
  • Column Name of diagnosisData file, column patient_id:  patient_id
  • Column Name of diagnosisData file, column admission_id:  admission_id
  • Column Name of diagnosisData file, column diagnosis_code:  diagnosis_code

Diagnosis Description input settings

  • Data full path:  /home/ronzano/Eclipse_WS/comorbidity4j-github/example/input/
  • DiagnosisData file name:  diagnosisDescription.csv
  • Column Name of diagnosisDescription file, column diagnosis_code:  diagnosis_code
  • Column Name of diagnosisDescription file, column diagnosis_description:  diagnosis_description

Disease Pairing settings

  • Data full path:  /home/ronzano/Eclipse_WS/comorbidity4j-github/example/input/
  • DiseasePairing file name:  diagnosisPairing.csv
  • Column Name of DiseasePairing file, column diagnosis_code:  diagnosis_code
  • Column Name of DiseasePairing file, column paired_diseases:  paired_diseases
  • Column Name of DiseasePairing file, column diagnosis_group:  diagnosis_group

Other parameters

  • Date format:  yyyy-MM-dd
  • Spreadsheet column separator char:  TAB
  • Enable multithreading:  true
  • Disease pairing default behaviour:  INDEX_DISEASES
  • Patient age computation approach (used if the patient age filter is enabled):  LAST_DIAGNOSTIC
  • P-value adjust approach:  BONFERRONI
  • Female identifier to compute sex ratio value:  Female
  • Male identifier to compute sex ratio value:  Male

Filters used to generate the analysis

  • Patient filter:  ---
  • Time directionality filter:  ComorbidityDirectionalityFilter [minNumDays=14]
  • Scores filter:  ---
Processing log
Below you can find the log of the comorbidity data analysis: 
exec_555399466 
exec_555399466 **************************************************
exec_555399466 ***** LOADING DATA FROM Disease pairing file...
exec_555399466 ***** Loaded from the Disease pairing file 6 diseases to consider of which:
exec_555399466 ***** - 4 diseases will be analized for comorbidity against all the diseases of the dataset;
exec_555399466 ***** - 2 diseases will be analized for comorbidity against all the diseases of the Disease pairing file (6 diseases);
exec_555399466 ***** - 0 diseases will be analized for comorbidity against a user defined list of deisease codes.
exec_555399466 **************************************************
exec_555399466 
exec_555399466 **************************************************
exec_555399466 ***** LOADING DATA FROM Patient file...
exec_555399466 ***** Patient file DATA LOADED: 100000
exec_555399466 **************************************************
exec_555399466 
exec_555399466 **************************************************
exec_555399466 ***** LOADING DATA FROM Admission file...
exec_555399466 ***** Admission file DATA LOADED: 361760
exec_555399466 **************************************************
exec_555399466 
exec_555399466 **************************************************
exec_555399466 ***** LOADING DATA FROM Diagnosis file...
exec_555399466 ***** Diagnosis file DATA LOADED for 361760 admission (with at least one diagnosis code)
exec_555399466 **************************************************
exec_555399466 
exec_555399466 **************************************************
exec_555399466 ***** LOADING DATA FROM Disease Description file...
exec_555399466 ***** Description loaded for 2604 diseases to consider of which:
exec_555399466 **************************************************
exec_555399466 **************************************************
exec_555399466 ***** Creating comorbidity dataset...
exec_555399466 OUTPUT OF COMORBIDITY DATASET CHECKS:

>>>>>>>>>>>>>>>>><<<<<<<<<<<<<<<<
>>> Comorbidity dataset stats <<<
    *****************************************************************************************************************************************
    *** IMPORTANT: these stats are computed WITHOUT applying any filter, considering the row data imported from the input files / dataset ***
    *****************************************************************************************************************************************
    > Num patients (WITHOUT filters applied - row data): 100000
    > Num visits (WITHOUT filters applied - row data):   361760
    > Num diseases (WITHOUT filters applied - row data): 2604.
    > 
    > Total visits WITHOUT diagnosis codes associated (over the whole NOT-filtered dataset): 0 over 361760 visits.
    > Total patients WITHOUT visits associated (over the whole NOT-filtered dataset): 0 over 100000 patients.
    > Total visits NOT associated to any patient (over the whole NOT-filtered dataset): 0 over 361760 visits.
    > 
    > DATA ERROR CHECK: No patients have one or more visits that occurred before the birth date.
    > 
    > Patient age statistics: age at first admission date: Stats{count=100000, mean=22.49515000000003, populationStandardDeviation=2.867876649631226, min=18.0, max=28.0}
    > Patient age statistics: age at first diagnosis date: Stats{count=100000, mean=22.49515000000003, populationStandardDeviation=2.867876649631226, min=18.0, max=28.0}
    > Patient age statistics: age at last admission date: Stats{count=100000, mean=52.32507999999955, populationStandardDeviation=17.257469773799425, min=18.0, max=92.0}
    > Patient age statistics: age at last diagnosis date: Stats{count=100000, mean=52.32507999999955, populationStandardDeviation=17.257469773799425, min=18.0, max=92.0}
    > Patient age statistics: age at execution date: Stats{count=100000, mean=60.29670000000015, populationStandardDeviation=17.26288067241386, min=28.0, max=98.0}
    > 
    > Most frequent (max 15) / less frequent (max 15) diseases with respect to the percentage of patients (WITHOUT filters applied - row data): 
    >       1 - 'F31' = 1227 patients (1.2270% of patients)  > Description: 9-group:  - Bipolar disorder, current episode hypomanic - Bipolar disorder, current episode manic without psychotic features - Bipolar disorder, current episode depressed, severe, without psychotic features - Bipolar disorder, current episode depressed, severe, with psychotic features - Bipolar disorder, current episode manic severe with psychotic features - Bipolar disorder, current episode depressed, mild or moderate severity - Bipolar disorder, current episode mixed - Bipolar disorder, currently in remission - Bipolar disorder
         > in Disease Pairing file, paired with the diseases 'true' for comorbidity analysis
    >       2 - 'F32' = 946 patients (.9460% of patients)  > Description: Major depressive disorder, single episode
         > in Disease Pairing file, paired with the diseases 'true' for comorbidity analysis
    >       3 - 'C90' = 675 patients (.6750% of patients)  > Description: Multiple myeloma and malignant plasma cell neoplasms
         > in Disease Pairing file, paired with the diseases 'true' for comorbidity analysis
    >       4 - 'D33' = 530 patients (.5300% of patients)  > Description: 4-group:  - Benign neoplasm of brain, supratentorial - Benign neoplasm of brain, infratentorial - Benign neoplasm of spinal cord - Benign neoplasm of cranial nerves
         > in Disease Pairing file, paired with the diseases 'true' for comorbidity analysis
    >       5 - 'C47.12' = 183 patients (.1830% of patients)  > Description: Malignant neoplasm of peripheral nerves of left upper limb, including shoulder
    >       6 - 'M05.42' = 180 patients (.1800% of patients)  > Description: Rheumatoid myopathy with rheumatoid arthritis of elbow
    >       7 - 'E10.65' = 177 patients (.1770% of patients)  > Description: Type 1 diabetes mellitus with hyperglycemia
    >       8 - 'C10.8' = 176 patients (.1760% of patients)  > Description: Malignant neoplasm of overlapping sites of oropharynx
    >       9 - 'F14.180' = 175 patients (.1750% of patients)  > Description: Cocaine abuse with cocaine-induced anxiety disorder
    >       10 - 'M06.02' = 175 patients (.1750% of patients)  > Description: Rheumatoid arthritis without rheumatoid factor, elbow
    >       11 - 'C57.02' = 174 patients (.1740% of patients)  > Description: Malignant neoplasm of left fallopian tube
    >       12 - 'D65' = 173 patients (.1730% of patients)  > Description: Disseminated intravascular coagulation [defibrination syndrome]
    >       13 - 'C49.5' = 172 patients (.1720% of patients)  > Description: Malignant neoplasm of connective and soft tissue of pelvis
    >       14 - 'C80.2' = 170 patients (.1700% of patients)  > Description: Malignant neoplasm associated with transplanted organ
    >       15 - 'Z12' = 170 patients (.1700% of patients)  > Description: Encounter for screening for malignant neoplasms
    >       2588 - 'B40.0' = 107 patients (.1070% of patients)  > Description: Acute pulmonary blastomycosis
    >       2589 - 'O99.355' = 106 patients (.1060% of patients)  > Description: Diseases of the nervous system complicating the puerperium
    >       2590 - 'E71.51' = 105 patients (.1050% of patients)  > Description: Disorders of peroxisome biogenesis
    >       2591 - 'D16.1' = 105 patients (.1050% of patients)  > Description: Benign neoplasm of short bones of upper limb
    >       2592 - 'B96.82' = 105 patients (.1050% of patients)  > Description: Vibrio vulnificus as the cause of diseases classified elsewhere
    >       2593 - 'C47' = 104 patients (.1040% of patients)  > Description: Malignant neoplasm of peripheral nerves and autonomic nervous system
    >       2594 - 'F94.2' = 104 patients (.1040% of patients)  > Description: Disinhibited attachment disorder of childhood
    >       2595 - 'D31.01' = 103 patients (.1030% of patients)  > Description: Benign neoplasm of right conjunctiva
    >       2596 - 'O9A.12' = 103 patients (.1030% of patients)  > Description: Malignant neoplasm complicating childbirth
    >       2597 - 'E10.22' = 102 patients (.1020% of patients)  > Description: Type 1 diabetes mellitus with diabetic chronic kidney disease
    >       2598 - 'M05.111' = 101 patients (.1010% of patients)  > Description: Rheumatoid lung disease with rheumatoid arthritis of right shoulder
    >       2599 - 'A42.0' = 101 patients (.1010% of patients)  > Description: Pulmonary actinomycosis
    >       2600 - 'C40.21' = 99 patients (.0990% of patients)  > Description: Malignant neoplasm of long bones of right lower limb
    >       2601 - 'F95.2' = 98 patients (.0980% of patients)  > Description: Tourette's disorder
    >       2602 - 'C04.0' = 97 patients (.0970% of patients)  > Description: Malignant neoplasm of anterior floor of mouth
    >       2603 - 'Z98.62' = 80 patients (.0800% of patients)  > Description: Peripheral vascular angioplasty status
    >       2604 - 'A18.7' = 62 patients (.0620% of patients)  > Description: Tuberculosis of adrenal glands    > 
    > 
    > Disease pairing to create disease pairs to analyze for comorbidity:
    >    Loaded 6 diseases read from the Disease Pairing file of which:
    >        - 4 diseases will be analized for comorbidity against all the diseases of the dataset;
    >        - 2 diseases will be analized for comorbidity against all the diseases of the Disease pairing file (6 diseases);
    >        - 0 diseases will be analized for comorbidity against a user defined list of deisease codes.
    > 
    > Number of disease codes included in the Disease Description file: 2604.
    > 
>>>>>>>>>>>>>>>>><<<<<<<<<<<<<<<<


exec_555399466 **************************************************
exec_555399466 ***** Info on the pairs of disases to study for comorbidity analysis...
exec_555399466 ***** Number of diseases pairs to study 20814.
exec_555399466 ***** Number of diseases of the dataset that are included in at least one disease pair to study 2604 (over 2604 diseases in input data).
exec_555399466 ***** 
exec_555399466 ***** Top-5 and botom-5 diseases paired with more / less diseases to study comorbidities:
exec_555399466 *****        > Disease 1 over 2604 participating in more pairs > DISEASE ID F44.4 (90) participates in 2603 disease pairs to study for comorbidity.
exec_555399466 *****        > Disease 2 over 2604 participating in more pairs > DISEASE ID C90 (94) participates in 2603 disease pairs to study for comorbidity.
exec_555399466 *****        > Disease 3 over 2604 participating in more pairs > DISEASE ID D33 (252) participates in 2603 disease pairs to study for comorbidity.
exec_555399466 *****        > Disease 4 over 2604 participating in more pairs > DISEASE ID I70.1 (1160) participates in 2603 disease pairs to study for comorbidity.
exec_555399466 *****        > Disease 5 over 2604 participating in more pairs > DISEASE ID F32 (214) participates in 5 disease pairs to study for comorbidity.
exec_555399466 *****        > Disease 2599 over 2604 participating in more pairs > DISEASE ID H36 (2599) participates in 4 disease pairs to study for comorbidity.
exec_555399466 *****        > Disease 2600 over 2604 participating in more pairs > DISEASE ID M06.251 (2600) participates in 4 disease pairs to study for comorbidity.
exec_555399466 *****        > Disease 2601 over 2604 participating in more pairs > DISEASE ID D37.0 (2601) participates in 4 disease pairs to study for comorbidity.
exec_555399466 *****        > Disease 2602 over 2604 participating in more pairs > DISEASE ID D40.12 (2602) participates in 4 disease pairs to study for comorbidity.
exec_555399466 *****        > Disease 2603 over 2604 participating in more pairs > DISEASE ID G47.21 (2603) participates in 4 disease pairs to study for comorbidity.
exec_555399466 *****        > Disease 2604 over 2604 participating in more pairs > DISEASE ID E87.4 (2604) participates in 4 disease pairs to study for comorbidity.
exec_555399466 ***** Comorbidity dataset created, including  2604 diseases.
exec_555399466 **************************************************
exec_555399466
Patient data overview
Below you can find a collection of charts useful to explore the features of the group of patient data you provided as input for comorbidity analysis. To explore the results of the comorbidity analysis over this set of patients, refer to the complete list of comorbidity pairs.
Important: you defined the following set of criteria to select / filter patients:  --- for comorbidity analysis.
The charts shown below provide an overview of all the patient data of your input dataset / files WITHOUT applying any filter!
Patients by age charts
Patients by birth-date charts
Patients by disease charts
Visits by disease charts
Sex ratio analysis
The following table shows the sex ratio of all pairs of diseases analyzed.
BA Sex Ratio column: given all individuals with disease A: (i) > 0 means prevalence of disease B in females; (ii) < 0 means prevalence of disease B in males; (iii) close to 0 means disease B is equally likely for females and males.
AB Sex Ratio column: given all individuals with disease B: (i) > 0 means prevalence of disease A in females; (ii) < 0 means prevalence of disease A in males; (iii) close to 0 means disease A is equally likely for females and males.
Comorbidity list
The following table shows the results of the comorbidity analysis: each row describe a pair of diseases.
The input dataset has been filtered by means of the following criteria (if any):
  • Patient filter:  ---
  • Time directionality filter:  ComorbidityDirectionalityFilter [minNumDays=14]
  • Scores filter:  ---
resulting in - disease pairs shown in the following table and evaluated with respect to the relevance of their comorbidity (see Input parameter section to review the input data used).
You can click on the header of each column to order disease pairs with respect to a specific comorbidity score. It is possible to resize columns as well as to change the order of each column by clicking on its header and draggin it.
You're considering both FEMALE and MALE
Interactive visualizations
From this tab you can interactively define criteria to further filter the complete list of comorbidity pairs.
The filtered set of comorbidity pairs will be visualized by means of a table and a heatmap to visually suport data exploration and analysis.
You're considering both FEMALE and MALE
Due to the high number of pairs that match filters (X - > 500 - pairs over X), your visualization could be difficult to explore and slow to generate.
We suggest to click on the 'Set more restrictive filters' button, change you filter setting and trgger data visualization again.

-

Interactive filter setting

Click on the checkbox to enable / disable each filter, then modify the filter parameters by clicking on the pencil icon.
Change filter P-value cut-off:
Current selection:
Value range:
Change filter Adj. p-value cut-off:
Current selection:
Value range:
Change filter Rel. risk cut-off:
Current selection:
Value range:
Change filter Phi cut-off:
Current selection:
Value range:
Change filter Comor. score cut-off:
Current selection:
Value range:
Change filter Odds ratio cut-off:
Current selection:
Value range:
Change filter Num. of patients cut-off:
Current selection:
Value range:
Pairs with disease: 

Parameter to visualize in the heatmap:

Results:

Please, properly set filters and other visualization parameters in the 'Interactive filter setting' box, then click 'Analyze!'
List of applied filters: No filter applied No filter applied No filter applied No filter applied No filter applied No filter applied No filter applied No filter applied
You're considering both FEMALE and MALE
Table of filtered comorbidities:
Time directionality is considered in the analysis of comorbidities
For each comorbidity pair shown in the table, the first temporal occurring disease is the disease A, followed in time by the disease B.

Heatmap visualization of filtered comorbidities:
Time directionality is considered in the analysis of comorbidities
For each comorbidity pair shown in the heatmap, consider as first temporal occurring disease the one on the y-axis (disease A), followed in time by the corresponding disease on the x-axis (disease B).


Network visualization of filtered comorbidities
(zoom with mouse scroll wheel and drag & drop nodes to explore the network):
Time directionality is considered in the analysis of comorbidities
The arrow of the ark that connects a pair of diseases represents their time directionality (the arrow points to the second disease occurring in time).


Heatmap visualization of sex ratio of the filtered disease pairs:
For each comorbidity pair shown in the heatmap, a sex-ratio value close to zero indicates that the co-occurrence of the disease on the x-axis, given a patient suffering the corresponding disease on the y-axis, is equally likely for males and females. A positive (negative) value of sex ratio indicate that the co-occurrence of the disease on the x-axis, given a patient suffering the corresponding disease on the y-axis, is more likely for females (males). Thus if the sex ratio has positive values the diagnosis of the x-axis disease in patients that have been diagnosed with the y-axis disease is more likely in females than males.

Info and contact
This analysis and visualizations have been performed by relying on Comorbidity4j, a java library useful to carry out comorbidity analyses. Detailed documentation and practical examples are available online at:
http://comorbidity4j.readthedocs.io/


Comorbidity4j is developed and maintained by the:
Integrative Biomedical Informatics Group
part of the Research Programme on Biomedical Informatics (GRIB), a joint research programme of the Hospital del Mar Medical Research Institute (IMIM) and the Department of Experimental and Health Sciences of the Universitat Pompeu Fabra in Barcelona.

If you need any support in using the tool or if you want to provide us with feedback and suggestions, please send an email to francesco<DOT>ronzano<AT>upf<DOT>edu.